0.49 Release Plan Discussion
Hey All,
It seems like it would be jumping the gun to put forth a proposed release plan without getting input from everyone on what should and should not be pursued during this cycle. After some feedback it should be a bit easier to put forth a somewhat realistic timeframe. Again, the goal is still to have a release before the end of the year. (Note: 0.48.4 is going to be cut hopefully in the next week or so)
First, I want to address, what major features and issues are still left that people want include and will be able to wrap up in the near future?
We still have some outstanding bugs with the new renderer which should really be addressed when possible. I know I will be going through the tracker soon to look for our blockers and must-fixes... others are welcome to do this as well.
Our "start" time is still pretty unfriendly... valgrind just screams that it's fontconfig taking up the biggest chunk of this. I have no clue though if anyone is interested in taking this on.
Also, it's time we get the skeletal pieces in place is the release notes. If you don't have time to write the full text, please write a placeholder Todo. Anyone is welcome to do this, if you are the author of a new feature or notable change or bugfix, traditionally you are the one to write this (as you're usually the most familiar with the feature). I don't believe we should include compatibility to compile against GTK3 as it will be viewed as a supported feature.
So, where do we stand. I would really like to get a proposed release plan out in the next week if possible.
Any suggestions for a theme for this release's About screen contest?
Cheers, Josh
One thing we really need to sort out is our approach to packaging, we still dont have a mac package for the last release... Having a hard timescale of when we announce post finalizing might be a good idea so we dont end up in limbo again. (even if it comes with an apology for certain platforms that the package is still being worked on) Trying to avoid Xmas would be good too, can we aim for before the end of Nov instead of before the end of the year?
Might be a good idea to have another irc meeting to discuss this issue (or a G+ hangout?)
Cheers
John
On Fri, Aug 17, 2012 at 8:23 AM, Josh Andler <scislac@...400...> wrote:
Hey All,
It seems like it would be jumping the gun to put forth a proposed release plan without getting input from everyone on what should and should not be pursued during this cycle. After some feedback it should be a bit easier to put forth a somewhat realistic timeframe. Again, the goal is still to have a release before the end of the year. (Note: 0.48.4 is going to be cut hopefully in the next week or so)
First, I want to address, what major features and issues are still left that people want include and will be able to wrap up in the near future?
We still have some outstanding bugs with the new renderer which should really be addressed when possible. I know I will be going through the tracker soon to look for our blockers and must-fixes... others are welcome to do this as well.
Our "start" time is still pretty unfriendly... valgrind just screams that it's fontconfig taking up the biggest chunk of this. I have no clue though if anyone is interested in taking this on.
Also, it's time we get the skeletal pieces in place is the release notes. If you don't have time to write the full text, please write a placeholder Todo. Anyone is welcome to do this, if you are the author of a new feature or notable change or bugfix, traditionally you are the one to write this (as you're usually the most familiar with the feature). I don't believe we should include compatibility to compile against GTK3 as it will be viewed as a supported feature.
So, where do we stand. I would really like to get a proposed release plan out in the next week if possible.
Any suggestions for a theme for this release's About screen contest?
Cheers, Josh
Live Security Virtual Conference Exclusive live event will cover all the ways today's security and threat landscape has changed and how IT managers can respond. Discussions will include endpoint security, mobile security and the latest in malware threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/ _______________________________________________ Inkscape-devel mailing list Inkscape-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/inkscape-devel
Hey John,
Thanks for responding! (shame, shame everyone else ;) ) No more playing friendly... packaging limitations won't matter. If we have a tarball, I don't care what we have packaged at announcement time. We'll do it like other projects and say "hey, it'll be available when/if someone packages it".
Pre-holiday was my goal... I'd like to see first or second week in Dec. A bonus to this is people can install fancy new stuff when their families get together. ;)
I think an IRC meeting is a great idea... Hangout has limitations for number of participants as well as is a less "privacy-oriented" way to participate. We still have people who participate that don't want to do so outright publicly.
Cheers, Josh
On Fri, Aug 17, 2012 at 3:11 AM, John Cliff <john.cliff@...400...> wrote:
One thing we really need to sort out is our approach to packaging, we still dont have a mac package for the last release... Having a hard timescale of when we announce post finalizing might be a good idea so we dont end up in limbo again. (even if it comes with an apology for certain platforms that the package is still being worked on) Trying to avoid Xmas would be good too, can we aim for before the end of Nov instead of before the end of the year?
Might be a good idea to have another irc meeting to discuss this issue (or a G+ hangout?)
Cheers
John
On Fri, Aug 17, 2012 at 8:23 AM, Josh Andler <scislac@...400...> wrote:
Hey All,
It seems like it would be jumping the gun to put forth a proposed release plan without getting input from everyone on what should and should not be pursued during this cycle. After some feedback it should be a bit easier to put forth a somewhat realistic timeframe. Again, the goal is still to have a release before the end of the year. (Note: 0.48.4 is going to be cut hopefully in the next week or so)
First, I want to address, what major features and issues are still left that people want include and will be able to wrap up in the near future?
We still have some outstanding bugs with the new renderer which should really be addressed when possible. I know I will be going through the tracker soon to look for our blockers and must-fixes... others are welcome to do this as well.
Our "start" time is still pretty unfriendly... valgrind just screams that it's fontconfig taking up the biggest chunk of this. I have no clue though if anyone is interested in taking this on.
Also, it's time we get the skeletal pieces in place is the release notes. If you don't have time to write the full text, please write a placeholder Todo. Anyone is welcome to do this, if you are the author of a new feature or notable change or bugfix, traditionally you are the one to write this (as you're usually the most familiar with the feature). I don't believe we should include compatibility to compile against GTK3 as it will be viewed as a supported feature.
So, where do we stand. I would really like to get a proposed release plan out in the next week if possible.
Any suggestions for a theme for this release's About screen contest?
Cheers, Josh
Live Security Virtual Conference Exclusive live event will cover all the ways today's security and threat landscape has changed and how IT managers can respond. Discussions will include endpoint security, mobile security and the latest in malware threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/ _______________________________________________ Inkscape-devel mailing list Inkscape-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/inkscape-devel
On 17/08/2012 09:23, Josh Andler wrote:
It seems like it would be jumping the gun to put forth a proposed release plan without getting input from everyone on what should and should not be pursued during this cycle. After some feedback it should be a bit easier to put forth a somewhat realistic timeframe. Again, the goal is still to have a release before the end of the year. (Note: 0.48.4 is going to be cut hopefully in the next week or so)
(…)
We still have some outstanding bugs with the new renderer which should really be addressed when possible. I know I will be going through the tracker soon to look for our blockers and must-fixes... others are welcome to do this as well.
WRT bug tracker: we haven't yet been adding the tag 'blocker' to confirmed regressions in current trunk (a few of them have temporarily been set to target milestone 0.49 to indicate that they ought be be fixed before releasing - this milestone target will be removed again once fixed). If helpful, I could compile a list of the most noticeable regressions to ease further discussing which of them ought to block the initial release (@JazzyNico - maybe we could join forces? ;) ).
Note: some regressions (after the merges of cairo-renderer & rendering cache and the removal of libnr) may only have been reported to the list here and not filed in the bug tracker - going through the related old threads, retesting the mentioned issues and filing reports for those not yet tracked at launchpad has been on my todo list for quite some time now (and this new thread might just be what makes me stop procrastinating it any further).
~suv
2012/8/17 Josh Andler <scislac@...400...>:
We still have some outstanding bugs with the new renderer which should really be addressed when possible. I know I will be going through the tracker soon to look for our blockers and must-fixes... others are welcome to do this as well.
The issues with the new renderer are as follows:
1. There are some rough edges caused by more SVG-compliant output (scaled strokes). More specifically, the selection cue no longer matches the visual bounding box of some objects.
2. The intermediate surfaces for pattern rendering are not computed correctly and the result is pixelation at high zoom levels. The relevant code is in sp-pattern.cpp lines 604-720 if someone wants to have a look.
3. Caching problems. In some instances, the caching mechanism does not invalidate everything it should. Sometimes it does not kep track of memory use properly and can take up several GB. There are some bug reports related to this. A related problem is that the cache is not preserved when objects are translated in the selector tool. Although this is only a missing optimization, it would substantially improve workflows involving blurred objects.
4. Cairo does not downscale bitmaps correctly. This is the worst issue, since it makes Inkscape useless when working with high resolution photos. The best solution is to implement mipmapping or at least better downscaling in Cairo. A temporary Inkscape-only solution would be to downscale the image to the correct size ourselves before outputting it to the screen.
Regards, Krzysztof
On Fri, Aug 17, 2012 at 11:23 AM, Josh Andler wrote:
Any suggestions for a theme for this release's About screen contest?
http://en.wikipedia.org/wiki/The_Wayfarer
:)
Alexandre Prokoudine http://libregraphicsworld.org
participants (5)
-
Alexandre Prokoudine -
John Cliff -
Josh Andler -
Krzysztof Kosiński -
~suv